WHAT ARE CHARACTERISTICS OF
CONVERTIBLE BOND ISSUERS?
Louren van Garderen 10288015 BSc Economie en Bedrijfskunde Finance and Organization Supervised by Rob Sperna Weiland
Abstract
We estimate the influence of firm characteristics of firms’ propensity to use convertible debt as a way of financing. We discuss motives for firms to use convertible debt, and propose R&D intensity and financial distress as characteristics. Based on a linear regression, we find a statistically significant positive effect of R&D intensity, and a significant negative effect of financial distress. We conclude that convertible debt is used to solve agency problems and overcome information asymmetries induced by these characteristics. This is important as it provides empirical insight on methods used by firms to solve this kind of problems.
Contents
Introduction ... 2
Background information ... 3
What are convertible bonds? ... 3
Motives for the use of convertibles ... 4
Literature review ... 6
Model setup and data description ... 9
Results and analysis ... 11
Discussion and conclusion ... 12
References ... 14
Appendices ... 17
Statement of Originality
This document is written by Student Louren van Garderen who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
Introduction
The Financial Times reported that convertible bond issuance surged last year (Bolger, 2014). Convertible bonds provide the holder the right to convert the bond into a pre-determined amount of equity. Firms can also use debt and equity to satisfy their financing needs since convertible bonds are essentially straight debt with a call option on the firms’ stock. According to Modigliani and Miller (1958), the financial structure of the firm has no influence on firm value. What is then the reason to issue convertible debt?
In the literature, most attention has been on valuation models. Billingsley and Smith (1996) note that valuation issues and call policies are well discussed, while research on characteristics of firms that use convertibles is scarce. Previous research has found several reasons for firms to issue convertible debt. Stein (1992) finds that convertibles are a way of issuing delayed equity. This is relevant when information problems make a direct issue costly (Myers and Maluf, 1984). Convertibles are also able to control managers’ adverse incentives caused by conflicts of interest between claim holders (Green, 1984). Firm characteristics discussed in the literature include a high ratio of R&D to sales, high distress costs, and risky cash flows (Essig, 1991).
This paper will seek an answer to the question: “What are characteristics of convertible bond issuers?”. Data on listed firms in the United States is selected and a linear regression is performed to estimate the influence of firm characteristics on the relative amount of convertible debt in their capital structure. This research will add to the literature by providing empirical insight on theoretical motives. Furthermore, it provides information regarding firm characteristics and thereby finds firms that benefit from the use of convertible debt.
The paper is structured as follows: Section 2 discusses background information on convertibles and motives to issue convertible debt. Section 3 reviews related literature on characteristics of firms issuing convertibles. Section 4 discusses the model that is used to estimate the effect of characteristics. Section 5 presents the results of the model and section 6 concludes.
Background information What are convertible bonds?
Convertibles are bonds with the option to be converted into a predetermined amount of equity. They can be described as a combination of straight debt and a call option on the firms stock. Some convertibles restrict the conversion to a specific time period, use time varying call prices, are issued with a call provision or can be converted in equity of another firm.
The value of convertibles responds to changes in coupon rate, market interest rate and maturity like straight bonds. The embedded option is essentially a claim on upward stock returns. Therefore, the value of a convertible also depends on the volatility of the underlying stock, stock price and dividend policy changes (Black and Scholes, 1973).
Convertible bonds are convertible at the convenience of the holder, allowing him to convert at the expiration date. This provides the upward potential of stock returns without additional downward risk. Holders will only convert when the conversion value is higher than the principal as this maximizes their value. Upon conversion, the firm will issue the pre-determined amount of stock, diluting existing shareholders interest. If no conversion takes place, the holder will retrieve the principal at expiration and no equity is issued.
Some convertibles have a call provision, generally restricted to a specific timeframe. A call provision gives the issuing firm the right to call, much like callable bonds. When called, the holder decides whether he prefers to retrieve the principal or convert and receive the pre-determined amount of shares. Optimal call policy describes that firms call outstanding bonds when they can be refinanced favorably (Asquith and Mullins, 1991) or as soon as the conversion value is higher than the principal. This limits the claim of the bondholder on the upper tail of profits, while ensuring that conversion takes place (Ingersoll, 1977). If the firm calls the bond knowing its holders will convert, it’s called a forced conversion.
The U.S. convertible market is well developed with $ 224 billion convertible debt outstanding and over $ 40 billion new issues in 2014. The U.S. account for about 50% of the worldwide trade in convertible debt. In 2014, many large U.S. companies (like Tesla Motors and Twitter) issued convertible debt, showcasing its relevance in today’s markets (Cherney, 2014). Convertibles are mostly traded in the secondary market. In 2002 the FINRA1
introduced TRACE (Trade Reporting and Compliance Engine) in an effort to increase price transparency. TRACE reports daily trading volumes and prices of trades in corporate and agency bonds in the U.S. This makes the U.S. market interesting to research as data is readily available.
1
Motives for the use of convertibles
In the literature, several motives are given for the use of convertible debt. These reasons will be discussed in this chapter.
Tax difference. Firms paying corporate taxes enjoy a tax advantage on interest payments because interest costs are tax deductible, as opposed to dividend payments. Asquith and Mullins (1991) discuss why firms don’t force call outstanding convertibles at the earliest possibility, in accordance with the optimal call policy as mentioned before. They find, based on an empirical analysis, that this is mostly due to the tax advantage enjoyed.
Billingsley and Smith (1996) survey managers on the motivations for issuing convertible bonds. They find that 32% of managers consider common stock as the main alternative to convertibles. As convertibles are comparable to equity they can be used as an instrument to save taxes, while enjoying the benefits of an equity-like financing structure.
Interest rate. The embedded call option of convertibles always has a positive value to the holder. The convertible therefore is at least as valuable as straight debt with the same characteristics. Because of the risk-free upward potential, investors accept lower interest rates. Consequently, convertible debt is less expensive as a funding source compared to straight debt (Green, 1984). Bancel and Mittoo (2004) question managers about considerations when issuing convertible debt. The belief that convertible debt is less expensive is important to 72% of questioned managers.
Over-investment. Jensen (1986) describes free cash flows as cash flow in excess of the amount needed to fund all profitable investments2
. The over-investment problem predicts that managers prefer investing free cash flows in unprofitable investments to distributing them to shareholders. This is what Berk and DeMarzo (2007) describe as “empire building”, concluding that mangers prefer investments that increase firm size, even if unprofitable. Jensen (1986) confirms that managers have incentives to grow the firm beyond its optimal size to gain control over more assets. Firm size is also positively associated with management compensation (Murphy, 1985). Mayers (1998) finds that convertibles are able to solve the overinvestment problem by creating exclusive signaling equilibria.
Risk–shifting. Jensen and Meckling (1976) find that the use of convertible debt could control firms' adverse incentive to shift investment to higher risk projects (also called the “asset
2
substitution problem”). Managers have incentives to take excessive risks if the firm faces financial distress. In case of bankruptcy, equity holders lose their investment and bondholders retrieve only a share of their initial investment. Due to limited liability it is in the interest of shareholders to invest in risky investments. Debtholders bear the downward risk, while shareholders retrieve the majority of the upward potential (Berk and DeMarzo, 2007). Green (1984) finds that using debt and equity results in an over-investment in risky projects if insiders cannot credibly signal their future investments at the time of funding.
A firm with more outstanding convertible debt has less incentive to transfer value from debtholders to shareholders (Jensen and Meckling, 1976). Doing so would make the embedded call option on the share price become valuable. Debtholders would exercise their right to convert. This conversion is a transfer of value from shareholders to debtholders. Spring (2002) agrees that convertible debt mitigates debt-based agency costs like risk-shifting since it exhibits characteristics of both debt and equity.
Risk estimation. Asymmetric information exists when issuing new financing. Issuers have incentives to undervalue the risk and overvalue investment possibilities. This results in adverse selection and induces the lemons’ premium discussed by Akerlof (1970). The high level of risk perceived by the market makes debt or equity funding unattractive. Convertibles are able to create a revealing equilibrium, solving the lemons’ problem (Brennan and Kraus, 1987). This finding is supported by the questionnaire of Bancel and Mittoo (2004), in which 55% of managers report choosing convertibles as a good way to signal about future growth opportunities.
Under-investment. When a firm issues new shares, the market values these securities at the expected value. Consider the situation where managers have private information about the firm’s future prospects. Due to information asymmetries and adverse selection, it is not possible to credibly inform the market (Myers and Majluf, 1984). The market will interpret the issue as bad news. Therefore, the share price is undervalued and a new issue of equity will result in dilution of current shareholders interest, resulting in less influence and a smaller claim on future profits.
Convertibles solve this problem because they are usually constructed so that conversion is only profitable when the stock price rises. This is a credible signal that the firm expects the share price to rise because managers prefer conversion (Brennan, 1986). The issue of convertible debt only issues equity trough conversion when share price has risen. Stein (1992) therefore calls convertibles ‘backdoor equity’. Managers consider this as one of the most important factors when issuing convertible debt (Bancel and Mittoo, 2004). According to their survey, 85% of managers expects the debt to be converted.
Literature review
In the literature, several characteristics of firms are discussed that could benefit from issuing convertibles. Mostly, these characteristics build upon agency or asymmetric information problems faced when issuing equity or debt as discussed before. In line with Stein (1992) we argue that the use of convertibles is highest amongst firms that have significant information problems or that face large financial distress costs when adding more debt. We use this to discuss two characteristics:
1. Research & Development. R&D has two characteristics that set it apart as an investment: uncertainty is high and it results in intangible assets, rather than tangible assets that provide insurance to the investor (Hall, 2002).
R&D investments are uncertain, especially at the beginning of a research program or project. In terms of financing, risk can be simply compensated for using a risk premium. Due to asymmetric information however, the market can’t easily value the risk which results in the lemons’ problem and underinvestment.
With asymmetric information, the firm has superior information. This is always the case with investments, at least to some extent. The market only observes highly aggregated financial indicators like sales or inventories on a quarterly or yearly basis, whereas insiders observe detailed information like daily sales or production levels per factory. But R&D is a major cause of information asymmetry. Aboody and Lev (2000) identify several reasons. First, R&D investments are relatively unique compared to tangible or financial assets. The latter can be sold more easily than the intangible assets generated by R&D. This uniqueness makes it harder to asses quality and value, as R&D investments can’t be easily compared to those of competitors. Secondly, there are no organized markets for R&D. This contributes to information asymmetry as investors can’t derive information from market prices and fluctuations. Financial or tangible assets in contrast, would reveal information. For example, the price of stock changes with firm value, as is influenced by efficiency and profitability. Thirdly, while financial statements provide information that could solve the information problem, they are of less value when assessing R&D. Firms are required to amortize and revaluate financial and tangible assets, but R&D is expensed as occurred. The market therefore can’t assess the real value of intangible assets created by R&D. Financial statements thus intensify the relative information asymmetries caused by R&D investments as compared to other investments.
Some empirical evidence for the relation between R&D and information asymmetries is found. For example, Barth, Kasznik, and McNichols (1998) find a significantly higher analyst
coverage for R&D intensive firms and a positive relation between the analyzing effort conducted by analysts and R&D intensity. They conclude that this is due to the investors’ information demand. Tasker (1998a) reports the same, as measured by a higher number of conference calls between the firm and analysts. Furthermore, Tasker (1998b) found that the majority of questions discussed in these conference calls are R&D related.
Information asymmetries could be resolved by supplying information to the market. However, moral hazard prevents the firm from directly transferring the information to lenders (Brealey, Leland and Pyle, 1977). Moreover, R&D intensive firms prefer not to disclose the nature of R&D investments to keep ahead of competitors (Bhattacharya and Ritter, 1983).
With asymmetrical information and no possibility to transfer information, signaling equilibria may not exist or may not be economically efficient. It is therefore impossible for the market to distinguish between good and bad investments. The lemons’ problem introduced by Akerlof (1970) argues that as markets can’t distinguish between investment quality, the market perceives the average quality. Therefore, good investments can’t be undertaken as the price of capital is based on the market average: good firms pay the lemons’ premium. Investments below average can be undertaken but result in losses for the investors. This induces a supply of low-quality firms and aggravates the lemons’ problem. If the asymmetric information is severe and supply of low-quality investments rises, the market for these investments might cease to exist.
Information transfer is necessary for good quality investments to be financed (Brealey, Leland, and Pyle, 1977). While firms can’t directly provide information, they can use actions to signal quality. Convertible debt is able to provide uniquely signaling equilibria, solving the risk estimation problem (Brennan and Kraus, 1987). Well-constructed convertible debt enables firms with good investments to send a signal. Without profitable investment opportunities, this would be too costly. The market interprets these signals accordingly, and the lemons’ premium diminishes or even disappears.
We conclude that convertible debt helps firms to solve information asymmetries that occur when issuing new financing. R&D intensive firms specifically face asymmetric information problems. We therefore expect convertible debt to be a relatively important way of financing for R&D investments.
2. Financial distress. Firms in financial distress generally find it hard to finance new investments. In particular, the threat of after-issue risk shifting becomes likely, making it costly to issue new debt. Convertibles are able to control the adverse incentives of risk-shifting induced by default risk (Stein, 1992).
Risk-shifting was first discussed by Jensen and Meckling (1976). It proposes that shareholders of a firm in distress could extract value from debtholders by increasing investment risk. Due to limited liability, shareholders receive at least zero. Therefore, further losses are absorbed by debtholders. As shareholders are the residual claimants on the upward potential, they benefit from increasing investment risk. At the time of a debt issue, debtholders are aware of this incentive and will require higher interests rates to compensate for the additional risk (Jensen and Meckling, 1976). High interest rates result in positive investments not undertaken, deteriorating firm value.
Firms with more outstanding convertible debt have less incentives to transfer value from debtholders to shareholders for several reasons (Jensen and Meckling, 1976). First of all, convertibles feature an embedded option. Black and Scholes (1973) make clear that option-holders benefit from higher volatility on the underlying asset. The embedded option makes the incentive effect of risk-shifting less severe as convertible bondholders share in the benefit from increased risk. This reduces shareholders profit from risk shifting and attenuates the losses of debtholders (Jensen and Meckling, 1976). Green (1984) confirms that the correct structure of convertible debt completely mitigates the risk-shifting problem. Secondly, convertible debt can credibly signal the availability of good investments, as discussed in the risk estimation motive. The signaling effect of convertible debt is especially severe for firms in financial distress, as these firms can’t carry the cost of more leverage and thus prefer to convert, thereby deleveraging the firm (Stein, 1992). Finally, another advantage of convertible debt is the lower required interest rate. As the embedded option always has a positive value (Black and Scholes, 1973), the required interest rate on convertible debt is lower than on straight debt (Green, 1984). The lower interest payments are especially important to firms in financial distress, considering they already find it difficult to cover interest payments.
Some previous empirical evidence is found by Mikkelson (1981), concluding that highly levered firms with a high growth rate (firms more likely to face financial distress) issue more convertible debt.
Therefore, we propose that firms facing financial distress use higher amounts of convertible debt in order to solve information problems, signal strength and mitigate risk-shifting. The next section presents our data and model setup.
Model setup and data description
We discussed R&D intensity and the presence of financial distress as characteristics, as convertible debt can be used to solve agency problems or diminish asymmetric information related to these characteristics. We use a linear regression to estimate the influence of these characteristics on firms’ propensity to rely on convertible debt as a way of financing.
In line with Stein (1992) we use the ratio of long-term convertible debt to total long-term debt as a proxy of a firms’ tendency to use convertible debt as a source of funding (‘relative convertible ratio’).
R&D intensity is measured as the ratio of R&D to sales, expressed in per cents (Stein, 1992, normalizing the R&D expenditures by firm size. Alternatively, the absolute amount of R&D expenditures is used, as the R&D to sales ratio is skewed by R&D intensive firms with a low level of sales, mostly firms still in development stages3
.
As in MacKie-Mason (1990), Altman’s (1968) Z-score model will be used to measure the presence of financial distress. The Z-score is a weighted average of five financial ratios. It is therefore readily available in the sample4
. Furthermore, we add a dummy variable for firms with an investment grade rating, based on Fitch Ratings’ long-term credit rating. The dummy equals one for ratings equal to BBB or better. Unfortunately, ratings where only available for 14% of observations. Therefore, a third measure of distress is added: credit spread on long-term debt. This is calculated as the difference between interest expenses divided by long-term debt outstanding and the U.S. Treasury Note interest rate5
. Almeida and Philippon (2007) find that, although the spread between government and corporate bonds is too large to be explained by distress costs alone, the spread can be used to estimate the distress risk adjustment. We do however note that credit spread is distorted to some extent as the interest rate is also influenced by the market interest rate at the time of issue, firm leverage, debt covenants, and more.
We control for industry factors by using the SIC industry sector identification and we use the natural logarithm of total assets to control for firm size. We also corrected for the leverage ratio6
, but this had no statistically significant influence on our results. This gives the following full regression model:
3
The R&D to Sales ratio can become extremely high when dividing by a small denominator. This supersedes our aim of normalizing the data. One such example in our dataset is ‘Northwest Biotherapeutics’ with $ 772.000 sales, $ 28.908.000 R&D expenditures and a R&D/Sales ratio of 3.744,5% in 2012, a firm with several cancer treatments in clinical trial (Northwest Biotherapeutics, 2015).
4
See appendix 3 for a brief description of the Altman Z-Score model.
5
The U.S. Treasury Note is used as it has a similar maturity to corporate long-term debt (10 years). The average interest rate in 2012 was 2.051% (TreasuryDirect, 2012).
𝑅𝑒𝑙𝑎𝑡𝑖𝑣𝑒 𝑐𝑜𝑛𝑣𝑒𝑟𝑡𝑖𝑏𝑙𝑒 𝑟𝑎𝑡𝑖𝑜 = ∝ + 𝛽𝑅&𝐷 𝑆𝑎𝑙𝑒𝑠∗
𝑅&𝐷
𝑠𝑎𝑙𝑒𝑠 + 𝛽𝑅&𝐷∗ 𝑅&𝐷 + 𝛽𝐴𝑡𝑙𝑚𝑎𝑛𝑍∗ 𝑍 + 𝛽𝐼𝑛𝑣𝐺𝑟𝑎𝑑𝑒∗ 𝐼𝑛𝑣𝐺𝑟𝑎𝑑𝑒 + 𝛽𝐶𝑟𝑒𝑑𝑖𝑡𝑆𝑝𝑟𝑒𝑎𝑑 ∗ 𝐶𝑟𝑒𝑑𝑖𝑡𝑆𝑝𝑟𝑒𝑎𝑑 + 𝛽𝐹𝑖𝑟𝑚𝑆𝑖𝑧𝑒∗ ln(𝑇𝑜𝑡𝑎𝑙𝐴𝑠𝑠𝑒𝑡𝑠) + 𝜖
We select firms listed on the Nasdaq or S&P500 in 2012 where data on outstanding straight and/or convertible debt use is available. We use the year 2012 because more recent years tend to show more ‘not available’ items. The average firm has $ 12.2 billion in assets, just over $ 2.8 billion of debt, of which $ 22.7 million convertible. The average Altman Z-score is above 5, signaling financial strength, and in line with industry estimates by Altman (2000). R&D/Sales are given for the year 2012 as well as the last 5 year. As most firms tend to spread out R&D investments (Hall, 2002), it is more appropriate to consider a long-term ratio instead of the 1-year average, which can be influenced by incidental changes in R&D or sales. Table 1 summarizes the sample.
The R&D/Sales ratio shows high variance and further analysis points out that this is due to a few very large observations. Therefore, we analyze the impact of R&D/Sales of firms with a ratio of 100% or less. This eliminates 78 observations. The Altman-Z score is based on several financial ratios. The disadvantage of ratios, just like R&D/Sales is their potential to get skewed by a small denominator. We normalize the Z-score by excluding values below -20 or above 20. We refer to appendices 1 and 2 where outlier selection is discussed of respectively R&D/Sales and Z-score.
Table 1 Summary statistics
Full sample Normalized sample
Mean
(Std. Deviation)
Obs. Mean
(Std. Deviation)
Obs.
Relative Convertible ratio .0277 (.1433) 2,904 .0329 (.1514) 934 R&D/Sales 367.5 (5429.4) 1,194 10.3138 (13.0344) 934 R&D Expenditures 220,809,000 (889,543,000) 1,627 337,828,000 (1,126,094,000) 934 Altman Z-score 5.14 (24.8) 2,128 3.9417 (4.7009) 934 Investment Grade 81,25% 352 81.81% 132
Results and analysis
Table 2 presents the results of the regression models. In model 1 we estimate the coefficients as proposed. The R&D to sales ratio is significant with a positive sign as expected. The other variables are not significant. The ‘Investment Grade’ dummy eliminates many observations, as only a few firms in the sample have a rating. Moreover, as it has a high standard error and seemingly no impact, the ‘Investment Grade’ variable is dropped in our further analysis. In model 2 the new observations introduce outliers7
. Therefore, we normalize regression 3 as explained before. We conclude that R&D/Sales ratio and Altman-Z are statistically significant at the 1% and 10%-level, respectively. For R&D and Credit spread remain insignificant, we leave those out in model 4. We find that explained variance is lower, but standard errors of remaining coefficients are lower as well. The Altman-Z Score is now statistically significant at the 5%-level. In model we estimate the influence of R&D expenditures on relative convertible use. We find no significant effect. When normalizing the expenditures using the natural logarithm (model 6), the coefficient is significant at the 1%-level. This suggests that the relation between R&D expenditures and the use of convertible debt is non-linear. We conclude that the normalized level of R&D is positively related to the use of convertibles.
In our further analysis, we focus on model 4. We conclude that the R&D to sales ratio is significantly positive at the 1%-level. A change in R&D to Sales ratio of 1% results in a 0.00169%-point higher Relative Convertible Ratio. Except for model 2, all estimations are statistically significant. R&D expenditures show mixed signs and no statistically significant results. The natural logarithm of R&D expenditures is significant. We therefore conclude that R&D has a statistically significant robust effect when normalized.
We find that the Altman Z-score coefficient is negative and statistically significant up to the 5% level. The sign is as expected, as a lower score means more distress. Credit spread is insignificant in all models. This is probably due to the fact that it is a rough estimate of financial distress. Unfortunately, this leaves us with only the Altman Z-Score in estimating the influence of distress. Although we find a statistically significant effect as expected, it is not robust.
7
Table 2 Results linear regression models (standard errors in parentheses) Model (1) (2) (3) (4) (5) (6) R&D/Sales 5 year 1.12E-03*** (4.16E-04) 2.85E-06 (1.83E-06) 3.47E-03*** (6.70E-04) 1.69E-03*** (3.88E-04) R&D -1.85E-09 (1.65E-09) 9.89E-10 (7.80E-09) -9.90E-09 (6.92E-09) 9.08E-10 (4.96E-09) LN R&D 1.99E-02*** (6.09E-03) Altman Z-score -7.20E-04 (1.21E-03) -1.17E-03 (8.98E-04) -3.41E-03* (1.75E-03) -2.52E-03** (1.05E-03) -1.54E-03** (8.13E-04) -3.16E-04* (1.75E-03) Investment grade -2.44E-05 (4.60E-03) Credit Spread 1.45E-07 (1.87E-06) -5.92E-06 (1.92E-05) -3.84E-06 (1.64E-05) 5.34E-07 (1.86E-05) LN Total Assets 2.38E-03 (2.08E-03) -2.62E-03 (3.77E-03) 4.04E-03 (2.16E-03) 4.04E-03 (2.16E-03) 4.28E-04 (1.92E-03) -156E-02*** 5.80E-03 R&D/Sales normalized no no yes yes no no Altman-Z normalized
no no yes yes yes yes
Obs. 122 542 568 934 1288 514
R2
0.0803 0.0588 0.0103 0.0281 0.0028 0.0290
Adj. R2
0.0323 0.0500 0.0019 0.0249 0.0005 0.0214
*,**,***: Significant at the 10%, 5%, and 1% level respectively.
Discussion and conclusion
Convertibles provide the holder the right to convert it into a pre-determined amount of equity. Firms can also use a sequential issue of debt and equity to satisfy their financing needs without influencing firm value. We are therefore interested in the characteristics of firms issuing convertible debt. Based on previous literature, we propose two characteristics that influence firms’ propensity to use convertible debt, R&D intensity and financial distress. We argue that convertible debt solves asymmetric information surrounding R&D investments. Furthermore, convertible debt is able to diminish adverse incentives and credibly signal investment opportunities in financial distress.
A linear regression is performed to estimate the impact of these characteristics. We found a statistically significant positive relationship between R&D intensity and the relative use of convertible debt. We therefore conclude that firms use convertible debt to solve asymmetric information problems surrounding R&D investments. Furthermore, we found suggestive
evidence that financial distress is negatively related with firms’ use of convertible debt. Although the Altman Z-Score is statistically significant, the robustness couldn’t be confirmed as neither the credit rating nor the credit spread has a significant impact. We conclude that firms tend to use convertible debt to mitigate adverse incentives in case of probable distress.
We’d like to note that the elimination of outliers not only adds explanatory power to our model, but also adds a selection bias, as firms with high levels of R&D are excluded. Furthermore, we used data of firms listed on the two major U.S. stock exchanges. This may induce a selection bias, as small and medium-sized firms usually aren’t listed. We do correct for firm size, but we can’t correct for firm sizes beyond the selected scope. Finally, we use data of outstanding convertibles and financials in the year 2012. While our results are indicative, this doesn’t necessarily view the considerations at the time of issue.
Further analysis is recommended to focus on issues rather than outstanding amounts, combined with financials at the time of issue. Alternatively, other firm characteristics can be considered. Finally, other measures of financial distress should be considered as a check for robustness. This is important as it provides insight in methods used to overcome agency problems and asymmetric information surrounding the financing of investments.
References
Aboody, D., & Lev, B. (2000). Information asymmetry, R&D, and insider gains.The Journal of Finance, 2747-2766.
Akerlof, G. A. (1970). The market for "lemons": Quality uncertainty and the market mechanism. The quarterly journal of economics, 488-500.
Alderson, M. J., & Betker, B. L. (1996). Liquidation costs and accounting data.Financial Management, 25-36.
Almeida, H., & Philippon, T. (2007). The risk‐adjusted cost of financial distress.The Journal of Finance, 62(6), 2557-2586.
Altman, E. I. (1968). Financial ratios, discriminant analysis and the prediction of corporate bankruptcy. The Journal of finance, 23(4), 589-609.
Altman, E. I. (2000). Predicting financial distress of companies: revisiting the Z-score and ZETA models. Stern School of Business, New York University, 9-12.
Asquith, P., & Mullins, D. W. (1991). Convertible debt: Corporate call policy and voluntary conversion. The Journal of Finance, 46(4), 1273-1289.
Auerbach, A. J., & Poterba, J. M. (1987). Tax loss carryforwards and corporate tax incentives. In The effects of taxation on capital accumulation (pp. 305-342). University of Chicago Press. Bancel, F., & Mittoo, U. R. (2004). Why do European firms issue convertible debt?. European
Financial Management, 10(2), 339-373.
Barth, M. E., Kasznik, R., & McNichols, M. F. (2001). Analyst coverage and intangible assets. Journal of accounting research, 1-34.
Berk, J. B., & DeMarzo, P. M. (2007). Corporate finance. Pearson Education.
Bhattacharya, S., & Ritter, J. R. (1983). Innovation and communication: Signalling with partial disclosure. The Review of Economic Studies, 50(2), 331-346.
Billingsley, R. S., & Smith, D. M. (1996). Why do firms issue convertible debt?. Financial Management, 93-99.
Black, F., & Scholes, M. (1973). The pricing of options and corporate liabilities. The journal of political economy, 637-654.
Bolger, A. (2014, July 21). Equity rally fuels convertible bond sales. Financial Times. Retrieved June 10, 2015, from http://on.ft.com/Ws7ZIM
Brealey, R., Leland, H. E., & Pyle, D. H. (1977). Informational asymmetries, financial structure, and financial intermediation. The Journal of Finance, 32(2), 371-387.
Brennan, M. (1986). Costless financing policies under asymmetric information.University of British Columbia working paper.
Brennan, M. J., & Schwartz, E. S. (1988). The case for concertibles*. Journal of Applied Corporate Finance, 1(2), 55-64.
Brennan, M., & Kraus, A. (1987). Efficient financing under asymmetric information. The Journal of Finance, 42(5), 1225-1243.
Broman, K. (1963). The use of convertible subordinated debentures by industrial firms 1949-1959. Quarterly Review of Economics and Business, 3(1), 65-75.
Cherney, M. (2014, October 5). Convertible Bonds Take Off in Low-Yield Era. The Wall Street Journal. Retrieved June 14, 2015, from http://on.wsj.com/1w2aCy4
De Roon, F., & Veld, C. (1998). Announcement effects of convertible bond loans and warrant-bond loans: An empirical analysis for the Dutch market.Journal of Banking & Finance, 22(12), 1481-1506.
Essig, S. M. (1992). Convertible securities and capital structure determinants. University of Chicago.
Green, R. C. (1984). Investment incentives, debt, and warrants. Journal of financial Economics, 13(1), 115-136.
Houston Jr, A. L., & Houston, C. O. (1990). Financing with preferred stock. Financial Management, 42-54.
Ingersoll, J. E. (1977). A contingent-claims valuation of convertible securities. Journal of Financial Economics, 4(3), 289-321.
Jensen, M. C. (1986). Agency cost of free cash flow, corporate finance, and takeovers. Corporate Finance, and Takeovers. American Economic Review,76(2).
Jensen, M. C., & Meckling, W. H. (1976). Theory of the firm: Managerial behavior, agency costs and ownership structure. Journal of financial economics, 3(4), 305-360.
Korkeamaki, T. P., & Moore, W. T. (2004). Convertible bond design and capital investment: The role of call provisions. The Journal of Finance, 59(1), 391-405.
Lee, H. W., & Figlewicz, R. E. (1999). Characteristics of firms that issue convertible debt versus convertible preferred stock. The quarterly review of economics and finance, 39(4), 547-563. MacKie‐Mason, J. K. (1990). Do taxes affect corporate financing decisions?.The Journal of
finance, 45(5), 1471-1493.
Mayers, D. (1998). Why firms issue convertible bonds: the matching of financial and real investment options. Journal of Financial Economics, 47(1), 83-102.
Moerland, P. W. (1995). Alternative disciplinary mechanisms in different corporate systems. Journal of Economic Behavior & Organization, 26(1), 17-34.
Murphy, K. J. (1985). Corporate performance and managerial remuneration: An empirical analysis. Journal of accounting and economics, 7(1), 11-42.
Myers, S. C., & Majluf, N. S. (1984). Corporate financing and investment decisions when firms have information that investors do not have. Journal of financial economics, 13(2), 187-221. Noddings, T., Christoph, S. C., & Noddings, J. G. (2013). International Handbook of
Convertible Securities. Routledge.
Northwest Biotherapeutics. (2015). Investors & Media. Retrieved June 25, 2015, from http://www.nwbio.com/investors-media/
Stein, J. C. (1992). Convertible bonds as backdoor equity financing. Journal of Financial Economics, 32(1), 3-21.
Tasker, S. C. (1998a). Bridging the information gap: Quarterly conference calls as a medium for voluntary disclosure. Review of Accounting Studies, 3(1-2), 137-167.
Tasker, S. C. (1998b). Technology Company Conference Call: A Small Sample Study. Journal of Financial Statement Analysis, 4, 6-15.
TreasuryDirect. (2012). Average Historical Interest Rates. Retrieved June 26, 2015, from https://www.treasurydirect.gov/govt/rates/pd/avg/2012/2012.htm
Appendices 1. Outliers R&D/Sales 5 year
We normalize using an upper tail of 100% R&D to Sales ratio. In result, skewness dropped from 22.75 to 2.72. We refer to Figure 1 and Figure 2, showing a boxplot of our sample before and after normalization.
Figure 1 R&D/Sales before normalization
Figure 2 R&D/Sales after normalization
0 50,000 100000 150000
RDtoSales5yr
0 20 40 60 80 100
2. Outliers Altman Z-score
While the Z-score is readily available in our dataset, not all observations are realistic. Altman (1968) a score of 3.0 or higher as ‘save’. Observations fare beyond 3.0 are considered extreme. As firms below 1.8 are considered to be ‘in distress’, we exclude values far beyond as outliers. More specific, we include all observations between -20 and 20. In result, skewness dropped from -1.06 to -0.45 and kurtosis from 199.62 to 6.46. We refer to Figure 3 and Figure 4, showing a boxplot of our sample before and after normalization.
Figure 3 Boxplot Altman Z-Score before normalization
Figure 4 Boxplot Altman Z-Score after normalization
-500 0 500
AltmanZ
-20 -10 0 10 20
3. Altman Z-Score
In 1968, Altman used a discriminant analysis approach to estimate the influence of several financial ratios on bankruptcy. Altman found that the combined value of five ratios was able to predict bankruptcy with greater significance than a sequential analysis. The Z-Score is calculated as follows. Firms with a score below 1.81 are in the ‘distress zone’ and likely to go bankrupt. A score above 3.0 means the firm is unlikely to go bankrupt in the next two years. Values between 1.81 and 3.0 are the ‘grey zone’ (Altman, 1968).
Z = 1.2 𝑊𝑜𝑟𝑘𝑖𝑛𝑔 𝐶𝑎𝑝𝑖𝑡𝑎𝑙 𝑇𝑜𝑡𝑎𝑙 𝐴𝑠𝑠𝑒𝑡𝑠 + 1.4 𝑅𝑒𝑡𝑎𝑖𝑛𝑒𝑑 𝐸𝑎𝑟𝑛𝑖𝑛𝑔𝑠 𝑇𝑜𝑡𝑎𝑙 𝐴𝑠𝑠𝑒𝑡𝑠 + 3.3 𝐸𝑎𝑟𝑛𝑖𝑛𝑔𝑠 𝐵𝑒𝑓𝑜𝑟𝑒 𝐼𝑛𝑒𝑟𝑒𝑠𝑡 𝑎𝑛𝑑 𝑇𝑎𝑥𝑒𝑠 𝑇𝑜𝑡𝑎𝑙 𝐴𝑠𝑠𝑒𝑡𝑠 + 0.6 (𝑀𝑎𝑟𝑘𝑒𝑡 𝑉𝑎𝑙𝑢𝑒 𝑜𝑓 𝐸𝑞𝑢𝑖𝑡𝑦) 𝑇𝑜𝑡𝑎𝑙 𝐿𝑖𝑎𝑏𝑖𝑙𝑖𝑡𝑖𝑒𝑠 + .999 𝑆𝑎𝑙𝑒𝑠 𝑇𝑜𝑡𝑎𝑙 𝐴𝑠𝑠𝑒𝑡𝑠
According to Altman (1968), the model differentiated between bankruptcy correctly in 95% of all cases. In 2000, Altman revisited the model of 1968 and the ZETA model8
and found that, while the ZETA model is more accurate, the Z-score is an appropriate measure of financial distress. One year prior, it is able to correctly predict whether a firm goes bankrupt in 93,9% of all cases (Altman, 2000).
